How does the brain distinguish between behaviorally relevant sensory input and that caused by an animal's own behavior? It has long been thought that signals related to behavior may be used to predict and cancel out sensory responses to animals' own movements. However, the neural mechanisms underlying this process have been elusive. Evidence was recently uncovered that self-generated sounds are cancelled at the first stage of auditory processing in mammals ? the dorsal cochlear nucleus (DCN) ? but the circuit mechanisms are unknown. Insights may come from the distinctive circuitry of DCN, which has striking similarities to the cerebellum, including Purkinje-like cartwheel cells (CWCs) that massively integrate non-auditory, behavior- related inputs that are subject to synaptic plasticity. Similar cerebellum-like circuits in electric fish are known to use behavior-related signals to generate predictions of the sensory consequences of behavior. These predictions take the form of highly specific ?negative images? that cancel out responses to self-generated sensory input. Aim 1 will explore the role of sensory prediction in the auditory system by testing whether DCN uses negative images to cancel self-generated sounds. Aim 2 will elucidate the role of CWCs in sensory cancellation by selectively monitoring and optogenetically manipulating their activity in awake, behaving mice. CWCs are the most numerous inhibitory interneuron in DCN, but evidence suggests they do not contribute to the processing of external auditory stimuli. These experiments will be the first to test whether CWCs contribute to processing self-generated sounds. This work can contribute to the treatment and understanding of tinnitus, a common and sometimes debilitating disorder in which sound is persistently perceived that is not actually present. Tinnitus has been associated with aberrant synaptic plasticity, somatosensory integration, and neuronal hyperactivity in DCN. Exploring the normal function of DCN plasticity and somatosensory integration, as well as the role of CWCs ? which potently inhibit DCN cells that are hyperactive in tinnitus ? could yield important insights into the pathology of tinnitus.